ABSTRACT: Tissue diagnosis is critical during surgical removal of solid cancers for margin evaluation and optimal excision. Complete resection of the tumor is associated with an improved prognosis for almost all types of solid malignancies. However, the standard clinical approach for intraoperative assessment of extent of tumor excision, frozen section analysis, can be subjective, require on site specialty pathology expertise, is labor- and time- consuming, and exposes patients to increased risks of anesthesia and surgical site infection. Our team has reported the development of a highly innovative handheld and biocompatible mass spectrometry device, the ?MasSpec Pen?, for rapid and non-destructive diagnosis of ex vivo and in vivo tissues (Zhang et al, Science Translational Medicine, 2017, 9, eaan3968). The MasSpec Pen provides the unique and transformative ability to assess molecular predictors of disease state directly from tissue samples without causing harm or damage to the tissues and in seconds. We believe the MasSpec Pen offers substantial improvements over existing technologies and represents an advance that is so innovative that will become a disruptive leap forward for intraoperative use in cancer diagnosis and margin evaluation. In collaboration with surgeons and pathologists at the Baylor College of Medicine, we propose to further develop and validate the MasSpec Pen within the context of its intended intraoperative use in surgical margin evaluation. The objectives of this proposal include: Aim 1. Optimize the MasSpec Pen for breast and pancreatic cancer detection. The MasSpec Pen provides near real time molecular detection capabilities for cancer diagnosis along with operational features that are attractive for clinical use. Technical refinements will be pursued to improve performance for intraoperative use, and to further determine and develop detection capabilities at various tumor cell concentration and tissue depths (Aim 1a). Performance measures (e.g. sensitivity and reproducibility) will be systematically evaluated. Statistical classifiers will be expanded considering aspects of tumor heterogeneity, refined, and their performance (sensitivity, specificity and accuracy) further validated using independent sample sets (Aim 1b). Aim 2. Validate the MasSpec Pen for intraoperative cancer diagnosis and surgical margin evaluation. The demonstration that the MasSpec Pen allows in vivo cancer diagnosis and surgical margin evaluation has profound clinical implications. Technical validation of cancer detection capabilities and final tuning of the technology for in vivo use will be pursued during breast (Aim 2a) and pancreatic (Aim 2b) cancer surgeries at Baylor College of Medicine. Predictive diagnosis and surgical margin status will be compared to clinical results for the same patients using standard clinical approaches to evaluate its clinical usefulness for clinical use.